Objective: The aim of this study was to compare the use of periodontal probing and cone beam computed tomography (CBCT) images in the diagnosis of furcation involvement (FI) in patients with chronic generalized severe periodontitis. Method and Materials: Fifteen patients with chronic generalized severe periodontitis were included in this study. In total, 174 furcation sites (all in molar teeth) were analyzed. FI was assessed at three sites (buccal, mesiopalatal, and distopalatal) of maxillary molars, and at two sites (buccal and oral) of mandibular molars. FI was assessed both clinically (periodontal probing) and on CBCT images, using a dichotomous scale (present/absent). The agreement between clinical and CBCTbased findings was calculated. Results: FI were more often detected by means of CBCT than by means of clinical examination. Agreement between the evaluation methods was present in 46.9% of cases (63.3% in maxilla, 45.0% in mandible). FI detected clinically was confirmed by means of CBCT in 24% of the evaluated sites. The largest agreement (73.7%) in FI detection was found in the distopalatal maxillary sites between CBCT and clinical probing. The smallest agreement (36.6%) was found in the buccal sites of the mandibular molars, in which 63.3% of FI were detected using CBCT only, but not clinically. Conclusion: The number of FI detected by means of CBCT was larger than by means of periodontal probing. In those cases in which chronic generalized severe periodontitis is clinically diagnosed, and surgical treatment is necessary, CBCT may be suggested as an adjunct tool for FI assessment. Keywords: cone beam computed tomography, furcation involvement, periodontal disease